Effects of Self-explanations as Scaffolding Tool for Learning Computer Programming

This Research to Practice Full Paper explores students' self-explanations in the context of programming. Specifically, this paper explores the use of in-code comments as an approach to support students' learning process and development of abstraction skills in an introductory programming course at the undergraduate level. Computer programming is a difficult skill to learn by novices due to the complexity of multiple elements interacting with each other to produce a specific outcome. Providing worked-examples paired with an engaging pedagogical practice has demonstrated to be an effective strategy for novices to start learning complex topics such as computer programming. One of the strategies that can support the introduction of worked examples is the use of self-explanation activities. In the context of programming, using in-code comments as a way for students to self-explain programming code can support the integration of worked-examples to scaffold their learning process. In this study, students wrote comments to explain how worked examples that were provided completed a specific task. Their comments were scored using an assessment rubric to provide detailed feedback about regarding the quality of their comments and their understanding of the code beyond the line by line execution. The goal of this study is to explore whether students with prior exposure to computer programming generate better self-explanations, and the effect that the quality of the written explanations and students' prior programming experiences have on student overall performance in the introductory programming course. The implications of this study will contribute to a better understanding of effective practices to incorporate worked examples and self-explanation activities in the form of in-code comments for introductory programming courses.

I. Introduction

Teaching programming to novices is a challenging task. Bennedsen and Casper [1] found that only two in three students enrolled in a computer programming course successfully completes the course, a finding that Watson and Li [2] corroborated in 2014. Learning how to program can be a complex task [3, 4]. Novice programmers need to simultaneously consider several interacting elements such as the purpose of the program, how the computer works, the programming language syntax and semantics, and algorithm design [3]. When students have not developed the appropriate schemata that enable them to solve problems from scratch on their own, these interacting elements may overwhelm students, often resulting in a cognitive overload [5].